Research On Online Observation Technology Of Rotor Pole Position Of Permanent Magnet Synchronous Traction Motor

Due to their excellent performance,permanent magnet synchronous motors are gradually used in electric traction systems for rail vehicles.Vector control technology is often used in permanent magnet synchronous motors.The premise is that the rotor pole position is known,and the train needs to determine whether the wheels are idling or coasting in time when running.Therefore,the rotor pole position and speed of the permanent magnet synchronous traction motor are in the full speed range Testing is essential.The detection of the magnetic pole position can be divided into two types: the use of position sensors and no position sensor detection.In the electric traction of rail vehicles,rotary transformers are mostly used as position sensors.The premise of using position sensors is that the deviation angle between the motor zero position and the sensor zero position is known.The zero deviation angle of a motor that is newly installed or reloaded after maintenance is unknown,and the problem of online detection of the zero deviation angle needs to be solved.To this end,this paper studies the magnetic pole position observation method with and without position sensor,and discusses the method of applying the method to the position sensor fault diagnosis and status self-check.This paper takes the permanent magnet synchronous motor with position sensor in the electric traction system of rail vehicles as the research object,and studies the calibration method of the zero deviation angle of the position sensor under the condition that the motor is installed.In the process of the train being dragged by a traction motor other than the calibration to a certain speed running state,the back-EMF zero-crossing detection method is used,combined with the principle of this method,to analyze how it is implemented on the vehicle,and the simulation test is used to verify its feasibility.When the train is at a standstill,a method is proposed to detect the initial position of the rotor by high-frequency injection and compare it with the resolver data to obtain the deviation angle.The algorithm is written into the train TCU control program,and the deviation angle can be detected before the train runs.The method was simulated and verified,and the results show that the error is within the allowable range.The study uses a sensorless algorithm to observe the position and speed of the magnetic pole,and adds it as a redundant system to the magnetic pole position fault diagnosis program.It compares the results with the position sensor in real time.When the position sensor fails,the observation results can be put into vector control in.It is proposed to use the sliding film observer method to observe at high speed in the motor,design and improve the sliding film observer,and perform simulation experiments on the traditional and improved sliding film observers to verify that the improved observation results are better than the traditional ones.Is more precise.At low speed,the high-frequency signal injection method is used,and the simulation results of the rotating high-frequency injection and pulse high-frequency injection methods are compared and analyzed.The composite observer fused with the weightingalgorithm is applied to the urban track permanent magnetic traction system.Simulation experiments are carried out on the track permanent magnetic traction system to verify that the composite observer can accurately observe the position of the magnetic pole and the permanent magnetic traction system at full speed.Good and stable work.Finally,an experimental verification was carried out on the d SPACE real-time simulation platform.The experimental results proved that whether the position sensor or the position sensorless control algorithm can be used to realize the online observation of the magnetic pole position of the permanent magnet traction motor,the accuracy and real-time performance of the system are verified.